Pedestals for Run 3 (2003)
For every RHIC fill a set of 240 pedestals has been loaded to the off-line DB.
So far data from 46 RHIC fills for days 120-150 (start April 30, fill F3587) have been processed.
Procedure:
10K first events from one of the first large file in every RHIC fill are processed.
Gaussian is fitted for each tower for ADC below 100.
The mean and the sigma of the Gaussian are loaded to the off-line DB with the time stamp of the start of the given RHIC fill.
QA:
Run R4123133 has been used as a reference. The following spreadsheet lists all towers for which pedestal has moved by 1 ADC count or width has changed by 30%.
table in CSV format
QA Summary:
a single pedestal position or width 'glitch' : 30 towers out of 240*46
more glitches : 05TB05, 05TC06, 05TC09, 05TE01-3, 06TA07
one permanent change :06TA03, 07TB09, 08TE10
run R4124051 has pedestals of 1 channel width (RMS=0) for towers 05TD01 -07TC12
Programs used:
to sort data & generate histo: big_ped.sh , runList.dat
automatic QA checkRuns.C
loading to DB see EEMC logbook (paz)